Slashdot Mirror
Breathalyzer Source Code Ruling Upheld
dfn_deux writes "In a follow up to a 2005 story where Florida judge Doug Henderson ruled that breathalyzer evidence in more than 100 drunk driving cases would be inadmissible as evidence at trial, the Second District Court of Appeal and Circuit Court has ruled on Tuesday to uphold the 2005 ruling requiring the manufacturer of the Intoxilyzer 5000, Kentucky-based CMI Inc, to release source code for their breathalyzer equipment to be examined by witnesses for the defense of those standing trial with breathalyzer test result being used as evidence against them. '"The defendant's right to a fair trial outweighed the manufacturer's claim of a trade secret," Henderson said Tuesday. In response to the ruling defense attorney, Mark Lipinski, who represents seven defendants challenging the source codes, said the state likely will be forced to reduce charges — or drop the cases entirely.' ... What this really means is that outside corporations cannot sell equipment to the state of Florida and expect to hide the workings of their machine by saying they are trade secret. It means the state has to give full disclosure concerning important and critical aspects of the case."
Finally
Guess there are some judges out there that understand justice.
Someone is going to figure out how to file a defense involving the release of Microsoft Windows, I just know it.
Bearded Dragon
The courts got it right, this time. Yeah, sure, the whole argument is a no-brainer for anyone who thinks about it for more than 30 seconds, but the jurists weighing similar cases re. voting machine source-code seem to be struggling with it nonetheless.
There's a joke here about the importance of open-source breathalyzers vs. voting machines but I'm too full of outrage fatigue to make an effort at it.
As pesky as the Bill of Rights can be to swift justice and severe vengeance, I do remember something about the the accused having a right to face their accuser. If the Intoxilyzer uses buggy firmware that results in inaccurate readings, then the accused has every right to question it. The company that makes the Intoxilyzer must be held responsible for its actions as well. Someday, somewhere, some company will step up and say, "Yes, we knew our product was faulty. But we have shareholders who will sell our stock in a heartbeat if we miss our mark in any quarter."
Holding both the accuser and the accused responsible for their actions is what helps create a society based on the rule of law. Otherwise, we'd be a police state. As a practicing trial attorney I *much* prefer the former.
=Smidge=
Is it just my observation, or is eldavojohn an idiot?
I know the "related stories" says this too, but just to get the ball rolling:
What this really means is that outside corporations cannot sell equipment to the state of Florida and expect to hide the workings of their machine by saying they are trade secret. It means the state has to give full disclosure concerning important and critical aspects of the case."
So, will this mean voting machine source code will have to be disclosed?
Personally, I'm most surprised that:
a) Governments don't require source code disclosure, at least for purposes of review, when they ask for bids or shop for equipment/software,
b) It's so hard for them to find someone willing to meet a).
Information theory is life. The rest is just the KL divergence.
The thing I like about open source is that everyone can make the software better. Why a company wouldn't want to produce source code, is beyond me. I think they fear that they will be viewed as incompetent when several mistakes are found in their code, or that sneaky randomizer function call rears its ugly head.
In all seriousness, companies would do well to realize that open source increases revenues by enabling a larger FREE workforce to do your work for you. Put aside your griefs with secrecy, unless of course your code doesn't work, or you stole large chunks of the code, and fear the legal ramifications.
I'd hate to be in their position, either way.
The dangers of knowledge trigger emotional distress in human beings.
No. Did you bother reading the summary? The judge ruled the defendant in a criminal case has the right to review the source code of the machine that was used to convict him.. It's not like they ruled that CMI had to open source the thing. That seems pretty reasonable to me.
My blog
No. It means a bunch of drunk drivers will be on the streets free to run whoever they want over. Even if the prosecutors move to blood test, the defendants will require a medical doctor or lab technician to testify as to the nature of the exam and how the test works, and perhaps the manufacturer of the reagents used in the tests have to verify that they are what they are. Tons of money have to be spent by the DA's office, which means higher taxes. Meanwhile, prosecution of drunk driving will go down, and more drunk drivers will be on the streets.
Everyone wins!
But ask yourself: if a judge dismissed exculpatory DNA evidence because the defense didn't procure a geneticist/scientist to testify on the stand, what would you say?
A NYC lawyer blogs. http://www.chuangblog.com/
I will bet money on one of three outcomes:
1. Breathalyzers cease to be used.
2. The source code will be released and showed to have MAJOR flaws or an algorithm that is not scientific at all.
3. The source code will be suddenly patched and every system will be required to be updated. The "new" source code will be released. Prosecution rates plummet, for some "unknown" reason.
#fuckbeta #iamslashdot #dicemustdie
As much as I like hearing about cases of stickin' it to Da Man, I don't know that we should necessarily celebrate this decision quite so much...
All software contains bugs. The defense will find some, and even if they only affect accuracy at the 7th decimal point, the case will get thrown out by a jury based on reasonable doubt. And this doesn't apply just to the current case, but to nearly any legal case using machine-generated evidence. The court allows DNA evidence? How about the firmware in the sequencing machine? Drug test came back positive? Let's see how Agilent's HPLC code rounds in integration.
Now, in some cases (*cough* Diebold *cough*) we may have a valid gripe against a closed-source implementation. But in most cases... Not to make this a case of "for the children", but do you want drunks behind the wheel? Screw the children (calm down, Mr. Jackson, I didn't mean it like that), I don't want to DIAF because someone can't stop at two beers.
That is sage advice for those of us who live in states with implied consent laws.
Out of modpoints but really liked a post? 1BDkF6TtmmeZ3yqXbz9yhdYVqRYnwFoXDj
Maybe this depends what state you are in... In California, you sign a contract to get your drivers license that says they can breathalyze or blood test you for BAC on reasonable suspicion of drunk driving and you waive the right to refuse.
Be aware that in some states (I think NC is one) failure to take the breathalyzer test will loose you your drivers licence. The penalties are the same as for drunk driving. Consult a lawyer in your state or country before taking slashdot advice. Kurt
If you're pulled over and suspected of DUI, then don't take the damn test
In many states, refusal to take a breathalyzer test is legally presumed an admission of guilt. You'll have a much harder time reversing a conviction based on a refusal to take the test (to wit: voluntary admission of guilt without evidence thereof) than challenging the accuracy of the instruments used.
Good luck with that.
(BTW: the whole "release the source code" thing is more a rasing-the-stakes legal tactic than a legitimate questioning of the equipment involved. Are you REALLY ready to expend considerable resources to find vindicating flaws in a commercial product? You have to convince the prosecutor you will do it, and succeed, before he'll drop charges in favor of keeping that revenue path flowing.)
Can we get a "-1 Wrong" moderation option?
in the UK, the hand-held testers are indicative only. If you fail the test, you're taken to the station for a breath test on a seriously big machine (think old minicomputer sized) or have a blood test, taken by a doctor, for use in a subsequent court case.
I'm sure there are all kinds of health-and-safety, human-rights, and civil-liberties reasons why blood samples cannot be taken at the roadside by a police officer.
Still, it's a start. If Florida is now legally required to use machines that make the source available, then they will show up as there is now a market for them. Other state departments will quite possibly start to use these knowing full well that if the closed source ones were successfully challenged in Florida, then they could be in their jurisdiction too.
Plus it's just good publicity. It's finally a case where the public understood that "Wait, this magical doohickey has to have a way to figure out the data it provides . . . and if you can't tell us how then we can't rely on what it says.".
"People who think they know everything are very annoying to those of us who do."-Mark Twain
Don't even need all 49. Consider the California Emissions standard. Many companies produce ONLY Cali rated cars because it's cheaper than adjusting their assembly line and shipping procedures to make custom cars only for California.
Looking around on the internet, I only see something like 3 professional grade breathalyzers. At this moment, any company looking to do business in Florida has to disclose their source code*. If two companies don't, that leaves the remaining one with a monopoly in Florida - *ChaChing*.
They might do this in Florida, but what if you get three or four other states passing the same rules? The pressure mounts.
I'm all in favor of this measure. I'm strongly against DUI, but that's countered by my even stronger desire for accuracy and accountability in government, especially criminal matters. Of course, I'm also for NOT counting it as a DUI unless you're actually, driving. Sleeping in the backseat of a dead-cold car in the bar's parking lot with the keys in your jacket isn't DUI.
*Well, they don't strictly have to, but Florida departments would be idiots to buy machines from companies that won't, as they're inadmissable as evidence.
I don't read AC A human right
If you're pulled over and suspected of DUI, then don't take the damn test, beacuse the accuracy of the breathalyzers are questionable. Plain and simple.
Worst. Advice. Ever. Let me qualify that, if you're a first offender, it's the worst advice ever. First off, depending on what state you're pulled over in, the consequences of refusing the test are worse than your first dui arrest. Second, prosecutors are now using the fact that you refused the test against you as proof that you were intoxicated. If you have no had prior DUI arrests, you should almost always take the test. You can always fight it later on, but you'll almost surely loose your license if you refuse.
You'll have that sometimes...
In this situation either you are actually drunk while driving, in which case you are a complete jerk and deserve to be locked up, or you are not drunk, in which case you'd save yourself a lot of unnecessary hassle if you just take the test. So your advice is only useful to the guilty trying to evade justice. Therefore, you are a TERRORIST!
Seriously though, I'm sure there is a small probability of a breathalyzer malfunction but that applies to everything else in the world, and there is a way of dealing with that if it happens (as in this case, challenging the evidence and perhaps getting it dismissed in court, requesting a blood test etc) In any case if that is your concern, how do you explain refusing to take the blood test etc. Also, the whole thing about police trying to "bust as many as possible" doesn't make sense, unless you mean to actually catch as many drunk drivers as possible? Isn't that a good thing? Or do you mean the cops somehow rig breathalyzers to show alcohol levels that aren't there?
Negative moral value of force outweighs the positive value of good intentions.
remembering back to Driver's Ed (1979) I think it was said that the human body metabolizes about 1 drink per hour. So if it takes an hour to get a blood sample, a suspect could fail a breath test but pass a blood test just by metabolism. Would a court factor in the time between inital arrest and blood sample collection?
Yes, alcohol is metabolized at a fairly regular rate. Since the time you were pulled over is known, and the time the test was administered is known, when you combine that with the relatively high accuracy/precision of the blood test you can determine what the BAC was at the time the person was driving.
Out of modpoints but really liked a post? 1BDkF6TtmmeZ3yqXbz9yhdYVqRYnwFoXDj
What this really means is that outside corporations cannot sell equipment to the state of Florida and expect to hide the workings of their machine by saying they are trade secret
No, what it means is that corporations that sell equipment THAT PRODUCE EVIDENCE TO BE USED IN CRIMINAL CASES can't hide behind trade secret laws. It's a very narrow set of circumstances. If the machine isn't used to produce criminal evidence, it isn't affected. Things like radar guns and red light cameras could be affected by this ruling. General consumer products are not.
The breathalyzer is effectively acting as a witness against the defendant in a DUI case. The defendant has a CONSTITUTIONALLY GUARANTEED right to cross-examine witnesses and challenge their credibility and accuracy. In the case of a machine, this can include subjecting the machine's design to scrutiny by a defense expert.
Seems pretty open & shut to me: if they don't disclose the engineering data necessary to validate the accuracy of the machine, then the evidence produced by the machine is inadmissible.
Since DUI is based on specific blood alcohol levels, they would have to drop those charges and settle for something where they could get a conviction based solely on the arresting officer's eyewitness testimony (EG reckless driving or other specific moving violations).
Why is it that the proponents of "one nation under God" are so eager to get rid of "liberty and justice for all"?
What about providing proof that it was the version of the source code that was made available for review that was running on the machine at the time the test was administered?
You do not have the right to refuse the test, but you do have the right to insist on them taking a blood test rather than a urine or breathalyzer test, and you do have the right to refuse the roadside ballet they try to make you do. I don't drink, but if I did, I'd insist on a blood test for two good reasons: It is going to be the most accurate thing, and arranging for it takes longer, giving your liver more time to reduce your BAC.
That is sage advice for those of us who live in states with implied consent laws.
Either way you are going to lose your license. The question is would you rather lose it through a civil process at DMV or would you rather lose it through the courts and get the added "bonus" of a criminal conviction? The best solution is obviously not to drive drunk but every lawyer I've ever talked to says to refuse the breath test if you've been drinking.
I want peace on earth and goodwill toward man.
We are the United States Government! We don't do that sort of thing.
minnesota is actually involved in a similar case with the same company: http://wcco.com/crime/breathalyzer.lawsuit.minnesota.2.669505.html. the only difference is that in our case, the state dept of public safety claims that its part of the contract that the source code be made available to the state and cmi is still refusing to provide source.
I'll go down to Florida tonight, get smashed, walk up to the first cop I see and demand the test. Three short months later me and my chemistry degree have a competing product on the market.
If you walk up to a cop and he arrests you for DUI, you may have a decent false arrest case on your hands - Forget engineering a breathalyzer.
He's getting rather old, but he's a good mouse.
I'm looking forward to seeing the secret breathing pattern the developers hid in the code so they could pass it every time.
http://www.mhall119.com
Honestly, it would probably be easier just to write it yourself. It's not some super-top-secret bit of magic code that no one else could reproduce. Have you ever tried working with someone else's code, with know knowledge or insight into the project?
It's not pretty. And this isn't an open source project with a wiki and people contributing to documentation etc. It might have been this one guy who worked there 5 years ago and never made a single code comment.
The code that does the actual work (the calculations) is probably very small. Most of it is probably written to interface with the device. And unless you are getting their exact device -- or one with identical specifications -- then you're going to have to rewrite that anyway. And I suspect they DO have a patent on the device, even if you did somehow get the code.
"That which does not kill us makes us stranger." -Trevor Goodchild
The penalties are most definitely *not* the same. You DO lose your driver's license. You do NOT have a DUI on your criminal record, probation, fines, etc.
Always refuse.
I don't know much about breathalyzers besides what they do. Is the source code really the limiting factor to you making a competing line of breathalyzers? I would think the sensor that measures the alchohol on your breath would be the most expensive and most difficult to manufacture part of the whole thing. Perhaps wrongly I would have assumed the source code would be the third easiest thing to make, behind the case for the thing and the hose you breathe into. In other words, I would have assumed that the source code would have been nothing too secret, while the actual sensor was what they spent a lot of money developing.
Of course, if the source code were very simple, I guess the company probably would have released it rather than facing the 2 mil fines. Or maybe that's just typical corporate arrogance.
Can someone explain this to me (hopefully keeping in mind that I have no background in coding)?
Of course, I don't believe drinking and driving should really even be a crime to begin with. There's already laws against hitting someone while driving.
The point is to stop you from driving impaired, BEFORE you hit someone, so that you don't hit someone.
Sort of like why 'attempted murder' is illegal. So they can legally stop you before you succeed.
But please, let's not get drunk drivers off the road. We need as many as possible out and about to cull the herd.
OK, screw the "insensitive clod" thing, you heartless trolling fuck. My sister, a PhD chemist, was killed on her way to work when some ass-hole, shit-faced drunk before 8 AM, crossed the median and demolished her car with his F150. My brother-in-law, at 29, became a single father of 2 in the early hours the next morning. The fuck driving the truck recovered just fine. "Culling the herd" my ass.
Breathalyzers may be crap, I know very little about them and no longer drink, but there is no excuse for allowing drunk drivers on the road.
The parties are not the ones who get to see confidential business information. Their attorneys and experts do. For example, I just worked on a case where both sides had to produce competition-sensitive documents to the other side in discovery. These documents were clearly marked "CBI," and I would have been in very serious trouble if I had sent these to my client. If I had done so on purpose, I could possibly have been disbarred. So no, the criminals probably won't get to see this source code. Their attorneys will give it to their experts, and if they find something really useful, they will ask the judge if they can pretty please use it in open court after stripping away anything superfluous to the reason for which it is being used. That will not be enough for their clients to make a competing product.
Today's Sesame Street was brought to you by the number e.
The sensor isn't really anything special anymore. Alcohol gas sensors are commodity item and easily had on the parts market. Typically they just output some analog voltage that varies according to the level of alcohol detected. That voltage is subject to calibration and then runs through an ADC to the actual code that maps voltage directly to some magic BAC number. Up until the voltage output, it's all easily tested. The calibration instructions I'll assume are also easily provided by the manufacturer since they [hopefully] perform calibration or provide instructions on how police should calibrate it. After calibrations, all bets are off. After all, it's the BAC number the device spits out, not the voltage of a discrete sensor, that the law is written around.
Analysis of the source needs to confirm that the scale of the sensor matches the scale in the source code used to generate that magic number. Imagine if one of them was logarithmic and the other was linear... there would be a lot of innocent people falsely convinced (and, depending on which one, a lot of guilty people could have gotten off).
I agree that the software will tell part of the story on whether or not the device is accurate, but I'd be interested in examining the hardware too. It is easy to imagine that varying conditions (temp, humidity, altitude, exhaust, smoker lung, etc) could alter the operation of the hardware even before the software comes into play. How have these variables been neutralized? Casting doubt on the device would be easy.
Casting doubt is what the defense is interested in, but what the public should really be interested in is the test data (from an independent third party). Have they conducted appropriate tests across sufficient body types and environmental conditions? Lets see the results.
There are many factors that could alter the reading on a breathalyser. Our bodies get rid of alcohol in many different ways and at varying rates. Some people will naturally blow higher ratings than others. In Australia, after a breathalyzer is used to detect potential drunk drivers, the driver is taken into a "booze bus" or the nearest police station where a blood sample is taken and tested to give the true result. The breathalyzer result should NEVER used in court.
Do you have some sort of evidence to support the idea that people who drive while intoxicated are more likely to infringe copyrights than people who don't drive while intoxicated?
Have you driven a fnord... lately?
You must wait a little bit before using this resource; please try again later.
Driving with a .08 BAC doesn't mean you're actively trying to cause harm.
That's your opinion.
In my opinion getting yourself reaction and judgment impaired and then hopping into the drivers seat is actively trying to cause harm.
Perhaps it shouldn't actually be a 'criminal offense'. But driving is a privilege not a right, and if you think its ok to get wasted and drive around you should have that privilege revoked.
Now you might argue that 0.08BAC is too low and that it doesn't affect you or whatever, fine, we can have a debate about what the actual number should be. Although I think 0.08 is in the right ballpark, and there are a number of studies which have shown that as you get drunk your ability to accurately gauge how drunk you are goes down. So the people arguing they are just fine at X BAC are far more often than not straight up wrong.